Identifying Rebel Twitter Users using Semi-Supervised Machine Learning
Published:
Identifying Rebel Twitter Users using Semi-Supervised Machine Learning by Dr Muhammad Ali Masood (2021)
Miscreant users use social media for attaining their purposes like recruitment, fundrais- ing, and spreading propaganda. Among diverse types of miscreant social media users, majority of the researchers have focused on identifying users associated with terror- ism and limited research is conducted for identifying rebel users. The focus of this thesis is to identify rebel users on social media. As many rebel users share strong sentiments, therefore it is important to understand these sentiments. We propose a novel sentiment classifier named Context-aware Sliding Window (CSW), which uses clues from the past sentiments to classify the sentiments of the current social media post. We also develop a new temporally labeled sentiment dataset. The proposed sentiment classification approach along with other proposed techniques is then used in identifying rebel users. For this purpose, we propose a Supervised Rebel Iden- tification (SRI) framework which uses a novel directed user graph as a feature of the framework. We convert the user graph into graph embeddings which models the contexts of the users in lower dimensions. For evaluation, we develop a first multi- cultural and multiregional dataset representing rebel users affiliated with nine rebel movements across five countries. Finally we propose a semi-supervised rebel identifi- cation (SSRI) frameworkto identify new rebel users. The experimental results for all the proposed approaches show significant improvements over the baselines.